White Matter Integrity Abnormalities in 1-to 5-Year-Olds with Global Developmental Delays: A Diffusion Study Utilizing Tract-Based Spatial Statistics and ROI-Based Analysis

Abstract Objects We aimed to use diffusion kurtosis imaging (DKI) and its derived white matter integrity metrics to characterize white matter integrity in children with global developmental delay（GDD） and explore the relationship between diffusion parameters and development quotient（DQ）. Methods and Materials Thirty-four children with GDD and 44 healthy control children were included in this study, and MRI images were acquired. Diffusion parameters were preprocessed and obtained by Pydesigner and FSL software, including fractional anisotropy (FA), mean diffusivity (MD), axonal diffusivity (AD), radial diffusivity (RD), anisotropy (MO), mean kurtosis (MK), axial kurtosis (AK), kurtosis fractional anisotropy (FAK), radial kurtosis (RK), and WMTI parameters: the axial diffusivity of the extra-axonal space (EAS_AD), the radial diffusivity of the extra-axonal space (EAS_RD), the diffusivity of the intra-axonal space (IAS_DA), the axonal water fraction (AWF), and the tortuosity of the extra-axonal space (EAS_TORT). Afterward, the Tract-based spatial statistics analysis was used to compare the diffusion parameters of the whole brain. We also performed covariance analysis to compare the diffusion parameters in the genu, body, and splenium of the corpus callosum. Lastly, we used partial correlation analysis to assess the correlation between diffusion parameters and DQ. Results In the results of TBSS, we found that FA, AD, MO, MK, RK, and EAS_TORT were significantly reduced(P＜0.05); RD, AK, KFA, AWF, EAS_AD, and EAS_RD significantly increased in children with GDD compared to healthy controls (P<0.05). In the corpus callosum, FA, MO, IAS_DA, and EAS_TORT were significantly reduced(P＜0.05); AK, KFA, and AWF were significantly increased (P<0.05). In the correlation analysis, fine motor skills DQ was negatively correlated with AK and AWF(P<0.05) and positively correlated with MO(P<0.05). In addition, Adaptability DQ was negatively correlated with AWF(P<0.05). Conclusion DKI and its derived WMTI parameters can characterize the white matter integrity of children with GDD and have a certain correlation with the DQ, which helps understand the specific pathological changes inside and outside axons and assists clinical diagnosis.